© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7631
    Prev
    Next
    INDEX
    Explanations

    annoying things and aggravations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åį±éĻ©
    -0.10
    æĢª
    -0.10
    丢äºĨ
    -0.09
    æĥĺ
    -0.09
    æĢĸ
    -0.09
    å¨ĺ
    -0.09
    éĵ¤
    -0.09
    éĿŀæ³ķ
    -0.09
    伤å¿ĥ
    -0.08
    è½®å»ĵ
    -0.08
    POSITIVE LOGITS
    åĨĹ
    0.11
    毫æĹł
    0.10
     hor
    0.10
    令
    0.10
    çĥ¦
    0.10
    令人
    0.09
    VERBOSE
    0.09
     verbosity
    0.09
    verbosity
    0.09
    èĥĻ
    0.09
    Activations Density 0.096%

    No Known Activations